What does an editor do?

The Science Learning Hub Pokapū Akoranga Pūtaiao offers literally thousands of videos and resources to help educators and the public with science education. There’s a team of people behind all the Hub’s video content – and before the information reaches viewers, it all goes through an editor.

Tom – the Science Learning Hub’s video editor

See more

Tom Goulter, the Science Learning Hub’s video editor, is shown working with Motion and Final Cut Pro software as he puts the finishing touches on videos produced about Aotearoa New Zealand’s growing space industry.

Rights: The University of Waikato Te Whare Wānanga o Waikato

I’m Tom, main video editor with the Science Learning Hub since the site began in 2007. My job involves wrangling raw footage – sometimes hundreds of hours’ worth – into the bite-sized fixes of information you see on the site. Today, I’ll explain the process a clip goes through on its way to you.

It all starts with research

Every story begins as an idea. We get approached by scientists and science organisations (like Crown Research Institutes and universities) about exciting research or outreach. We’ll then identify what educators across Aotearoa need. Teachers will also let us know what the gaps are, and we reach out to our wide network of scientists for content as well. We then craft content that builds learners’ understanding of science ideas and concepts and the nature of science. Often we’ll plan a suite of content around a theme such as space or animal ethics. It can take a while to plan the content direction and organise interviews and camera technicians. During that time, we’ll research the topic and familiarise ourselves with the science.

Everyone’s familiar with filmed interview footage from documentaries or the news. It can be surprising to learn that, before asking the questions, an interviewer (typically our producer Rachel) has to familiarise themselves so well with the work that they could almost give their own answers!

Camera crew interview man outdoors

See more

Hub producer Rachel interviewing Mark Goodwin of Plant & Food Research in an avocado orchard. Once Rachel finished the interview, the camera team filmed the trees and nearby beehives for additional/cutaway footage.

Rights: The University of Waikato Te Whare Wānanga o Waikato 

When they’re conducting an interview, our team will typically shoot footage of the science being discussed or scientists going about their mahi. This might mean watching them in the lab or field or it can be as simple as filming them typing on a laptop. No matter the work, it all needs writing up!

The paper edit

Once we have the footage, the first task is to create a version with timecode – a numerical readout representing the place of every single frame within the footage. This version is used to generate a transcript. Both human transcribers and AI-based tools are used, though an AI-generated transcript will typically require considerable human input for it to be usable.

Paper edit example

See more

Snippets of interview dialogue are marked up and crossed out to show what to use in the edit. This image shows how the director and editor work together on a transcript in the early stages of producing a video.

Rights: The University of Waikato Te Whare Wānanga o Waikato

From this transcript – which includes timecoded references for every question, answer and shot – a paper edit is assembled. This is another text document showing which portions of the interview are to be used. I’ll assemble this paper edit using editing software, referring back to the timecoded transcript to pull the required words from the interview.

Written text is very different from spoken speech, and often a speaker’s words don’t sound quite as the transcript suggests. Our scientists and experts work at very high levels, and a big part of science communication is distilling their knowledge to be clearly communicated to a first-time viewer. I’ll remove the ‘ums’ and repetitions – little quirks of speech that we all have but don’t always notice. Working with the producer, we rearrange and shape the spoken-word track to be as clear and concise as possible.

Sometimes it’s necessary to script a voiceover to give context not provided by the interviewees. Some clips (like our Bird of the Year series) are scripted entirely by me and our team, whereas others will use voiceover to clarify what hasn’t been stated. Usually I’ll record this myself, but this is often what we call a ‘bash track’ – a non-final version used to get the timing right before we bring in a final voiceover artist.

Cutaways and images

Now comes the part where an edit really comes to life – adding cutaway footage and imagery. Remember earlier where we gathered footage of our scientists going about their mahi? We now return to that footage, cutting together sequences that tell a visual story to illustrate the spoken-word track.

Ideally, much of what we want to show has been shot as part of our source footage, but it’s almost inevitable that sooner or later we’ll run into a need for imagery that wasn’t shot on the day. This is where our library of pre-existing footage is a real gold mine. We might have shot something for a previous project that can be reused in a new story. Our most reused footage is from 2013 and shows cows going to the toilet!

Other sources for covering footage include Wikimedia Commons, YouTube clips licensed for reuse and paid sites like Getty Images and 123rf. Some cutaway footage is used to illustrate what’s being said, but most of the time it’s necessary to cover visually distracting edits to the spoken-word track. The best covering edits do both!

Animations and effects

Sometimes the ideal covering footage might depict something it’s not possible to film. The ideas or processes might contain abstract details we need to help viewers understand. That’s where animations or motion graphics can be useful. I’ve generated our own graphics depicting things like the Māori account of creation, the migration of the Pacific’s first settlers, and how satellites scan Aotearoa for methane emissions.

Think about a project like our Huia – Bird of the Century animation. The huia went extinct long before cameras could film it, so there’s no accessible videos or even photographs of a living bird for us to use here. We used archival drawings and taxidermy specimens to resurrect the bird – albeit to illustrate that technology can never fully bring back what humans have destroyed!

Sound and picture sweetening

Once the words and pictures are more or less where we want them to be, it’s time for audio mixing and colour grading.

Mixing involves going through the audio for each clip and adjusting the sound levels of every piece of video so the story is communicated clearly. Most important is getting the interview vocals clearly audible. I’ll boost the quiet parts, turn down the loud parts and add fades and filters to make the dialogue crisp and clear. Alongside this, almost every piece of video has one or more audio tracks, and we need to decide where to mix these in. Again, the aim is to help tell the story without distracting from the vocals. The best way to improve the picture, one adage goes, is to make the sound better!

Colour grading

See more

Editing software is used to adjust visual aspects. This is a shot with colour grading applied to just the left-hand side.

Rights:  Carol Brieseman/The University of Waikato Te Whare Wānanga o Waikato 

It’s also necessary to make the visuals look as nice as they can. Colour grading (sometimes called colour timing) is the process of going through shot by shot to make everything clear and eye-catching. No camera can capture visuals as clearly as the human eye sees them. Underexposed footage loses detail in the darkness, whereas overexposed shots blaze white-hot and obscure what we want to see. Footage shot indoors will have very different light to that captured outdoors, and even different lightbulbs have different colour temperatures that are far more noticeable on screen than to the naked eye. Colour grading can’t fix everything, but we hope to use the best parts of what’s been filmed and to grade each shot so it sits nicely in a sequence.

Output and upload

Getting a clip right requires ongoing back-and-forth between the editor, producer, the rest of our team and a story’s other stakeholders. Every piece of footage and audio needs to be accounted for, and even every person shown in the frame needs to have consented that we've represented their image and/or research fairly. This can take a number of revisions and replacements of this or that piece of footage before we arrive at an edit that everyone’s happy with.

Timecode markers in video footage

See more

This is a shot from a project in progress, marked with yellow timecode. Using timecode enables the team to make precise comments about feedback and edits.

Rights: Carol Brieseman/The University of Waikato Te Whare Wānanga o Waikato

While we’re still working on a clip, successive edits I share will once again have timecode laid over them. This makes them easier to comment and feed back on and clearly shows that the footage isn’t yet ready for the public. Once we’ve arrived at a final version (or a first final – which can be the first of many!), I output a version in higher resolution without the timecode. We’ll add this to our Vimeo library and I generate subtitles to help with accessibility, and it will slot into an article or page to provide added context for what’s seen.

It’s impossible to overstate the importance of crediting and attributing footage. The end of almost every clip on the Science Learning Hub contains credits and attributions showing where our vocals and visuals have come from. Collating this information isn’t glamorous but it ensures our materials will have longevity, ensuring accessibility and hopefully leading to many satisfied viewers.